A Numerical and Experimental Investigation of the Machinability of Elastomers

(1)

RODKWAN, SUPASIT. A Numerical and Experimental Investigation of the Machinability of Elastomers. (Under the direction of Dr. John S. Strenkowski.)

In this dissertation, the machinability of elastomers is investigated. The main objective of the

research is to determine the machining conditions for which an elastomer can be machined

with a smooth surface finish. Both machining experiments and numerical simulations were

carried out to achieve this goal. For the experimental studies, a series of orthogonal

machining tests were conducted to investigate the effects of various machining parameters on

chip morphology, machined surface condition, and resulting machining forces. Feed speed

and rake angle were found to have a significant effect on the type of chip generated during

orthogonal machining. High feed speed conditions and large rake angle tools produced long

and continuous ribbon-like chips and a corresponding smooth machined surface. The design

of the workpiece fixture in the lathe was also found to be critical for machining smooth

surfaces.

In the numerical investigation, half-wedge indentation and orthogonal cutting models were

developed to simulate incipient separation of elastomers. It was found that high tensile

normal stress and maximum principal stresses, as well as a large concentrated strain energy

density near the separation point lead to favorable conditions for the formation of continuous

chips and a good surface finish. The indentation simulations correlated well with the cutting

tests in which tools with a large rake angle and large feed produced continuous chips and a

smooth surface finish. The models offer potential for identifying the cutting conditions and

(2)

OF THE MACHINABILITY OF ELASTOMERS

by

SUPASIT RODKWAN

A dissertation submitted to the Graduate Faculty of

North Carolina State University

in partial fulfillment of the

requirements for the Degree of

Doctor of Philosophy

MECHANICAL AND AEROSPACE ENGINEERING

Raleigh

2002

APPROVED BY:

___________________________

(3)

Dedicated to my parents, my wife and…

(4)

BIOGRAPHY

Supasit Rodkwan was born in Bangkok, Thailand in 1969. He was an only child in

the family raised by his mother, Preeya Rodkwan and his father, Chid Rodkwan. He

graduated from Samsenwittayalai High School in Bangkok where his mother spent her

almost entire career as a teacher in social science and history. His father, who received his

Master degree in Public Administration from Syracuse University in 1972, has recently

retired as a vice-governor at Satool province in southern region of Thailand. In 1990, Mr.

Rodkwan received his Bachelor of Engineering degree in Mechanical Engineering from King

Mongkut University of Technology Thonburi in Thailand. Soon after his graduation, he

began his professional career as a mechanical engineer in a mechanical erection department

at TPI Polene Co., Ltd., one of the largest cement production companies in Thailand. In

1993, he came to pursue his graduate study at the University of Southern California where he

received his Master of Science in Mechanical Engineering with a concentration in Computer

Aided Design/Computer Aided Manufacturing/Computer Aided Engineering

(CAD/CAM/CAE) in 1995. Then, he enrolled as a graduate student in the Mechanical

Engineering Department at the University of Kansas where he later received his Master of

Science in Mechanical Engineering majoring in Computational Solid Mechanics in 1999. In

1997, he transferred to North Carolina State University to begin his Ph.D. study in an area of

computational mechanics applied to manufacturing processes with Dr. John S. Strenkowski.

After receiving his Ph.D. degree, he plans to return to Kasetsart University in Thailand where

he has been appointed as a lecturer in the Department of Mechanical Engineering since

(5)

ACKNOWLEDGEMENTS

I am very grateful to Dr. John S. Strenkowski, the chairman of my Ph.D. advisory

committee, for his invaluable support, guidance and endless encouragement since the

beginning of my enrollment at North Carolina State University. It has been my honor to

have an opportunity to work with him for a number of years. I consider myself to be very

fortunate to have him serving as my advisor. I also would like to thank Dr. John A. Bailey,

who taught me the fundamentals of machining and material deformation during my graduate

study, Dr. John S. Stewart, who provided me an valuable advice in machining and partial

financial assistance during my Ph.D. study, and Dr. Melur K. Ramasubramanian for his

kindness to serve on the Ph.D. advisory committee. In addition, I would like to thank Dr.

Albert J. Shih and Juan-Pablo Gallego for their valuable suggestions throughout the research

work in the machining of elastomers.

I also owe gratitude to Dr. Karan Surana, Deane E. Ackers Distinguished Professor of

Mechanical Engineering at the University of Kansas, who first introduced me to the world of

the finite element method. In addition, I would like to thank Dr. Yi-Min Wei and Dr.

Chang-Chi Hsieh, Dr. Xiangfu Wang, former graduate students in the Department of Mechanical

and Aerospace Engineering at North Carolina State University, for their valuable discussions

on the subject of the machining process. Appreciation is extended to Rufus (Skip)

Richardson and Mike Breedlove for their assistance in preparation of cutting tools,

specimens, and lathe setup. Special thanks are also given to my colleagues in the elastomer

(6)

I gratefully acknowledge the financial support from the National Science Foundation

as well as the industrial collaboration of Michelin Americas R&D Corporation and Lord

Corporation. Additionally, my graduate study at North Carolina State University was made

possible with partial funding from the Department of Mechanical and Aerospace

Engineering, the Integrated Manufacturing Systems Engineering Institute (IMSEI), and the

Wood Machining and Tooling Research Program (WMTRP) at North Carolina State

University. Their courtesies are always greatly appreciated.

Finally, I would like to express my deep appreciation to my parents, and my dearest

wife, Nussara Boonkungwan, for their unconditioned love, inspiration and support during an

endless period of my graduate study in the United States. I hope that completion of my Ph.D.

(7)

1. INTRODUCTION ... 1

1.1 Scope and Objective of Dissertation ...4

1.2 Overview of Dissertation ...6

2. LITERATURE REVIEW ... 7

2.1 General Description of Elastomers ...7

2.2 Analysis of the Strain Energy Function for Elastomers ...10

2.2.1 A Statistical Approach ... 10

2.2.2 A Phenomenological Approach... 11

2.3 Analysis of a Wedge Indentation Process ...14

2.3.1 Wedge Indentation of Metals ... 14

2.3.2 Wedge Indentation of Elastomers... 17

2.4 Analysis of a Machining Process ...19

3. ELASTOMER MACHINING EXPERIMENTS... 26

3.1 Introduction ...26

3.2 Experimental Setup ...26

3.3 Experimental Results and Discussions...31

4. DESCRIPTION OF MODEL DEVELOPMENT ... 46

4.1 A General Description of the ABAQUS Software...46

4.1.1 ABAQUS/Standard ... 48

4.1.2 ABAQUS/Explicit ... 49

(8)

TABLE OF CONTENTS (continued)

4.3 Contact and Interaction Concepts in Wedge Indentation and Orthogonal Cutting

Models...53

4.4 Description of Indentation and Orthogonal Cutting Models...56

4.4.1 Modeling of Indentation of AISI 4340 Steel ... 57

4.4.2 Modeling of Orthogonal Cutting of AISI 4340 Steel ... 61

4.4.3 Modeling of Indentation of Elastomers ... 67

4.4.4 Modeling of Orthogonal Cutting of Elastomers ... 75

5. SIMULATIONS OF ORTHOGONAL CUTTING AND WEDGE

INDENTATION... 80

5.1 Simulations of AISI 4340 Steel...80

5.1.1 Orthogonal Cutting Simulation ... 80

5.1.2 Wedge Indentation Simulation with a 70 degree Half-Wedge Angle .... 87

5.2 Simulations of Indentation of Elastomers...93

5.2.1 Wedge Indentation for a 80 degree Half-Wedge Indenter (Case 1)... 97

5.2.2 Wedge Indentation for a 40 degree Half-Wedge Indenter (Case 2)... 102

5.2.3 Wedge Indentation for a 40 degree Half-Wedge Indenter and Tip Edge Radius of 0.01270 mm (Case 3) ... 110

5.2.4 Wedge Indentation for a 40 degree Half-Wedge Indenter for Feed of 0.2540 mm (Case 4)... 117

5.2.5 Wedge Indentation for a 40 degree Half-Wedge Indenter and a Friction Coefficient of 1.20 (Case 5) ... 122

5.2.6 Wedge Indentation for a 40 degree Half-Wedge Indenter and Boundary Condition Type 2 (Case 6)... 127

(9)

TABLE OF CONTENTS (continued)

6. CONCLUSIONS AND FUTURE WORK... 140

6.1 Conclusions ...140

6.2 Future Work ...142

(10)

LIST OF TABLES

Table 2.1 Comparison of properties of elastomers with other materials ...9

Table 4.1 A summary of the material properties of AISI 4340 steel...60

Table 4.2 Temperature-dependent thermal expansion coefficients for AISI 4340 steel. ..63

Table 4.3 Temperature-dependent elastic properties of AISI 4340 steel...63

Table 4.4 Temperature-dependent elastic-plastic properties of AISI 4340 steel...64

Table 5.1 Summary of operating conditions used in wedge indentation simulations of

elastomers.. ...94

Table 5.2 Summary of operating conditions for evaluating various effects on elastomer

(11)

LIST OF FIGURES

Figure 2.1 Comparison between orhogonal and oblique cutting. ...20

Figure 2.2 Nomenclature of orthogonal cutting...20

Figure 3.1 Fixture design and cutting tool setup for the elastomer machining

experiments ...27

Figure 3.2 Side view of workpiece in the lathe chuck for the elastomer

machining experiments. ...28

Figure 3.3 Nomenclature of cutting tool angles...29

Figure 3.4 Chip types for various rake angles and feed speeds for a cutting speed of

2.50 m/s...32

Figure 3.5 Photographs of machined surfaces for a cutting speed of 2.50 m/s, three rake

angles and two feed speeds ...33

Figure 3.6 Typical measured machining forces in three axes (cutting, thrust, and

transverse components) for a sharp cutting tool with a rake angle of 10 degrees, feed

speed of 0.0635 mm, and cutting speed of 2.50 m/s...35

Figure 3.7 Typical measured machining forces in three axes (cutting, thrust, and

transverse components) for a sharp cutting tool with a rake angle of 50 degrees, feed

speed of 0.2540 mm, and cutting speed of 2.50 m/s...35

Figure 3.8 Measured force components for a sharp cutting tool with various tool rake

angles, a feed speed of 0.0635 mm and a cutting speed of 2.50 m/s ...37

Figure 3.9 Measured force components for a sharp cutting tool with various tool rake

angles, a feed speed of 0.2540 mm and a cutting speed of 2.50 m/s ...37

Figure 3.10 Comparison of measured cutting forces for cutting tools with various rake

angles and feed speeds for a cutting speed of 2.50 m/s ...38

Figure 3.11 Comparison of measured thrust forces for cutting tools with various rake

angles and feed speeds for a cutting speed of 2.50 m/s ...38

Figure 3.12 Apparatus for measuring the cutting edge sharpness ...39

Figure 3.13 Schematic diagram to determine the tool edge radius...40

Figure 3.14 Unused tool edge with a nose width of 10 microns (10 degree rake

(12)

LIST OF FIGURES (continued)

Figure 3.15 Used tool edge with nose a width of 15 microns (10 degree rake angle,

5 degree clearance angle)...43

Figure 3.16 Unused tool edge with a nose width of 10 microns (30 degree rake angle, 5 degree clearance angle) ...44

Figure 3.17 Used tool edge with nose a width of 15 microns (30 degree rake angle, 5 degree clearance angle)...44

Figure 3.18 Unused tool edge with a nose width of 10 microns (50 degree rake angle, 5 degree clearance angle) ...45

Figure 3.19 Used tool edge with nose a width of 15 microns (50 degree rake angle, 5 degree clearance angle)...45

Figure 4.1 ABAQUS modules ...47

Figure 4.2 Contact and interaction discretization ...54

Figure 4.3 Initial mesh of a half-wedge indentation model of AISI 4340 steel...58

Figure 4.4 Initial mesh of a half-wedge indentation model of AISI 4340 steel (detailed view) ...58

Figure 4.5 Initial mesh for the orthogonal cutting model of AISI 4340 steel...62

Figure 4.6 Initial mesh for half-wedge indentation model of elastomers ...68

Figure 4.7 Initial mesh for half-wedge indentation model of elastomers (detailed view). ...68

Figure 4.8 Uniaxial tensile stress-strain relationship for elastomer used in the wedge indentation model. ...69

Figure 4.9 Schematic of the uniaxial deformation mode ...71

Figure 4.10 Schematic of the equi-biaxial deformation mode...71

Figure 4.11 Schematic of the planar deformation mode...71

Figure 4.12 Initial mesh for the orthogonal cutting model of elastomers...76

Figure 4.13 Uniaxial tensile stress - extension ratio relationship used in orthogonal cutting model of elastomers...78

(13)

LIST OF FIGURES (continued)

Figure 5.2 Von Mises stress contours (MPa) at early configuration (analysis step 3,

increment 60) ...83

Figure 5.3 Von Mises stress contours (MPa) at final configuration (analysis step 4,

increment 6). ...83

Figure 5.4 Tresca stress contours (MPa) at early configuration (analysis step 3,

increment 60).. ...84

Figure 5.5 Tresca stress contours (MPa) at final configuration (analysis step 4,

increment 6). ...84

Figure 5.6 Normal stress contours (MPa) at early configuration (analysis step 2,

increment 11). ...85

Figure 5.7 Normal stress contours (MPa) at final configuration (analysis step 4,

increment 6)... ...85

Figure 5.8 Strain energy density contours (MPa) at early configuration (analysis

step 3, increment 60)...86

Figure 5.9 Strain energy density contours (MPa) at final configuration (analysis

step 4, increment 6)...86

Figure 5.10 Undeformed configuration of the finite element mesh for a half-wedge

indenter angle of 70 degrees ...87

Figure 5.11 Von Mises stress contours (MPa) at an indentation depth of 0.02 mm for a

half-wedge indenter angle of 70 degrees. ...89

Figure 5.12 Von Mises stress contours (MPa) at an indentation depth of 0.05 mm for a

half-wedge indenter angle of 70 degrees.. ...89

Figure 5.13 Tresca stress contours (MPa) at an indentation depth of 0.02 mm for a

half-wedge indenter angle of 70 degrees. ...90

Figure 5.14 Tresca stress contours (MPa) at an indentation depth of 0.05 mm for a

half-wedge indenter angle of 70 degrees.. ...90

Figure 5.15 Tresca stress contours (MPa) within a plastic shear zone of 478 MPa

(14)

LIST OF FIGURES (continued)

Figure 5.16 Tresca stress contours (MPa) within a plastic shear zone of 478 MPa

at an indentation depth of 0.05 mm for a half-wedge indenter angle of 70 degrees...91

Figure 5.17 Strain energy density contours (MPa) at an indentation depth of 0.02 mm

for a half-wedge indenter angle of 70 degrees...92

Figure 5.18 Strain energy density contours (MPa) at an indentation depth of 0.05 mm

for a half-wedge indenter angle of 70 degrees...92

Figure 5.19 Deformed and undeformed configurations of the finite element mesh for

a half-wedge angle of 80 degrees, tip edge radius of 0.00635 mm, feed of

0.0635 mm, friction coefficient of 0.60, and indentation depth of 0.10 mm. ...97

Figure 5.20 Tresca stress contours (MPa) for a half-wedge angle of 80 degrees,

tip edge radius of 0.00635 mm, feed of 0.0635 mm, friction coefficient

of 0.60, and indentation depth of 0.10 mm.. ...99

Figure 5.21 Detailed view of Tresca stress contours (MPa) for a half-wedge

angle of 80 degrees, tip edge radius of 0.00635 mm, feed of 0.0635 mm, friction

coefficient of 0.60, and indentation depth of 0.10 mm ...99

Figure 5.22 Normal stress contours (MPa) for a half-wedge angle of 80 degrees,

of 0.60, and indentation depth of 0.10 mm... ...100

Figure 5.23 Detailed view of normal stress contours (MPa) for a half-wedge

Figure 5.24 Detailed view of maximum principal stress contours (MPa) for a

half-wedge angle of 80 degrees, tip edge radius of 0.00635 mm, feed of 0.0635 mm,

friction coefficient of 0.60, and indentation depth of 0.10 mm ...101

Figure 5.25 Detailed view of strain energy density contours (MPa) for a half-wedge

(15)

LIST OF FIGURES (continued)

Figure 5.27 Tresca stress contours (MPa) for a half-wedge angle of 40 degrees,

tip edge radius of 0.00635 mm, feed of 0.0635 mm, friction coefficient of

0.60, and indentation depth of 0.10 mm.. ...104

Figure 5.28 Detailed view of Tresca stress contours (MPa) for a half-wedge angle

of 40 degrees, tip edge radius of 0.00635 mm, feed of 0.0635 mm, friction

of 0.60, and indentation depth of 0.10 mm.. ...105

coefficient of 0.60, and indentation depth of 0.10 mm. ...105

Figure 5.31 Detailed viewed of maximum principal stress contours (MPa) for a

friction coefficient of 0.60, and indentation depth of 0.10 m ...106

Figure 5.33 Normal stress of the elastomeric workpiece area under the indenter tip

for a tip edge radius of 0.00635 mm, feed of 0.0635 mm, and friction coefficient

of 0.60 with various half-wedge angles.. ...108

Figure 5.34 Maximum principal stress of the elastomeric workpiece area under the

indenter tip for a tip edge radius of 0.00635 mm, feed of 0.0635 mm, and friction

(16)

LIST OF FIGURES (continued)

Figure 5.35 Strain energy density of the elastomeric workpiece area under the indenter

tip for a tip edge radius of 0.00635 mm, feed of 0.0635 mm, and friction coefficient

of 0.60 with various half-wedge angles ...109

Figure 5.36 Deformed and undeformed configurations of the finite element mesh

for a half-wedge angle of 40 degrees, tip edge radius of 0.01270 mm, feed of

0.60, and indentation depth of 0.10 mm... ...112

Figure 5.38 Detailed view of normal stress contours (MPa) for a half-wedge angle

coefficient of 0.60, and indentation depth of 0.10 mm.. ...112

friction coefficient of 0.60, and indentation depth of 0.10 mm. ...113

Figure 5.41 Normal stress of the elastomeric workpiece area under the indenter tip

for a tip edge radius of 0.01270 mm, feed of 0.0635 mm, and friction coefficient

of 0.60 with various half-wedge angles ...115

Figure 5.42 Strain energy density of the elastomeric workpiece area under the

indenter tip for a tip edge radius of 0.01270 mm, feed of 0.0635 mm, and friction

coefficient of 0.60 with various half-wedge angles.. ...115

Figure 5.43 Effect of indenter tip edge radius on the strain energy density

and maximum principal stress of the elastomeric workpiece area under feed

of 0.0635 mm, and friction coefficient of 0.60 for various half-wedge angles

(17)

LIST OF FIGURES (continued)

0.60, and indentation depth of 0.10 mm ...119

coefficient of 0.60, and indentation depth of 0.10 mm... ...110

angle of 40 degrees, tip edge radius of 0.00635 mm, feed of 0.2540 mm,

friction coefficient of 0.60, and indentation depth of 0.10 mm.. ...120

Figure 5.49 Effect of indentation feed on the strain energy density and

maximum principal stress of the elastomer workpiece area under an indenter tip for a

tip edge radius of 0.00635 mm, and friction coefficient of 0.60 for various

half-wedge angles at 10 mm indentation depth.. ...121

tip edge radius of 0.00635 mm, feed of 0.0635 mm, friction coefficient of 1.20,

and indentation depth of 0.10 mm ...124

(18)

LIST OF FIGURES (continued)

Figure 5.55 Effect of friction on workpiece/indenter interface on the strain energy

density and maximum principal stress of the elastomeric workpiece area

under the indenter tip for a tip edge radius of 0.00635 mm, and feed of

0.0635 mm for various half-wedge angles at 10 mm indentation depth...126

Figure 5.56 Deformed and undeformed configurations of the finite element mesh

for a half-wedge angle of 40 degrees, tip edge radius of 0.00635 mm, feed

of 0.0635 mm, friction coefficient of 0.60, indentation depth of 0.10 mm

and a boundary condition Type 2...127

tip edge radius of 0.00635 mm, feed of 0.0635 mm, friction coefficient of 0.60,

indentation depth of 0.10 mm and a boundary condition Type 2.. ...129

coefficient of 0.60, indentation depth of 0.10 mm and a boundary condition Type 2...129

friction coefficient of 0.60, indentation depth of 0.10 mm and

a boundary condition Type 2 ...130

(19)

LIST OF FIGURES (continued)

Figure 5.61 Effect of fixture design of the workpiece on the strain energy density

and maximum principal stress of the elastomeric workpiece area under the indenter

tip for a half-wedge angle of 40 degrees, tip edge radius of 0.00635 mm,

feed of 0.0635 mm, friction coefficient of 0.60, and indentation depth of

0.10 mm for various half-wedge angles at 10 mm indentation depth...131

Figure 5.62 Undeformed and deformed configurations of the finite element mesh.. ...133

Figure 5.63 Von Mises stress contours (MPa) at the analysis step 3, increment 33...135

Figure 5.64 Von Mises stress contours (MPa) at the analysis step 3, increment 33

(detailed view)... ...135

Figure 5.65 Tresca stress contours (MPa) at the analysis step 3, increment 33.. ...136

Figure 5.66 Tresca stress contours (MPa) at the analysis step 3, increment 33

Figure 5.67 Normal stress contours (MPa) at the analysis step 3, increment 33.. ...137

Figure 5.68 Normal stress contours (MPa) at the analysis step 3, increment 33

Figure 5.69 Maximum principal stress contours (MPa) at the analysis step 3,

increment 33...138

Figure 5.70 Maximum principal stress contours (MPa) at the analysis step 3,

increment 33 (detailed view)... ...138

Figure 5.71 Strain energy density contours (MPa) at the analysis step 3, increment 33..139

Figure 5.72 Strain energy density contours (MPa) at the analysis step 3,

(20)

LIST OF SYMBOLS

W Strain energy density or free energy per unit volume

λ1, λ2, λ3 Principal stretch ratios along three perpendicular axes

N Number of chains per unit volume

k Boltzmann constant

T Absolute temperature

G Shear modulus or modulus of rigidity

C1, C2, C3 Material constants obtained from experiment

I1,I2, I3 Strain invariants

αk A real number

m A positive integer

µk, αk Materials constants obtained from experiment

A

Area of one fracture surface of a crack

c

T

Critical tearing energy

[M] Mass matrix

[C] Stiffness matrix

[K] Damping matrix

1 +

n

u

&&

Nodal accelerations

1

+

n

u& Nodal velocities

1

+

n

u Nodal displacements

a n

(21)

LIST OF SYMBOLS (continued)

t

∆ Time step

p

ε

_&

Effective plastic strain rate

ef

σ

Effective yield stress

Y

σ

Initial yield stress

D, p Material constants based on empirical strain rate sensitivity

F Deformation gradient

i

ε

Principal nominal strains

i

λ Principal stretches

U

λ Stretch in the loading direction

U

T Uniaxial nominal stress

B

T Equi-biaxial nominal stress

S

T Nominal planar stress

c

τ

Critical shear stress of the chip along the chip/tool interface

µ

Friction coefficient

th

τ

Threshold stress related to material failure

p

Normal presure at a contact point along the chip/tool interface

f

Critical shear function

n

(22)

LIST OF SYMBOLS (continued)

f

σ

Material failure stress under tensile mode

n

τ

Measure shear stress measured at a specifc distance

f

τ

Material failure stress under shear mode

(23)

1. INTRODUCTION

Elastomers are widely used in various industries for tires, springs, shock isolators, noise and

vibration absorbers, seals, corrosion and abrasion protection, and electrical and thermal

insulators. Among elastomeric materials, rubber is the most well known elastomer. Rubber

has been used as an engineering material since the mid 1800’s (Charlton and Yang, 1994).

Some of the unique properties that make elastomers different from other conventional

materials such as metals include the following:

1. Elastomers exhibit a highly non-linear stress-strain relationship as compared to metals.

2. Elastomers are able recover to an undeformed state after large deformation upon removal

of the load.

3. Elastomers have no yield point in the stress-strain curve and they are highly non-linear at

large strains. The elastic modulus and tensile strength of elastomers are also low

compared to metals.

4. Elastomers have good fatigue resistance and high energy absorption capacity due to their

high elasticity.

5. The material properties of elastomers change significantly with temperature variation.

6. Poisson’s ratio of elastomers is very close to 0.50 (nearly incompressible).

7. The coefficient of friction for elastomer surfaces is relatively high compared with a metal

to metal interface.

(24)

9. The moldability of elastomers allows for their use in a wide range of applications

such as hoses, belts, tubes, engine mounts, wiper blades, footwear, toys, and tires.

In comparison with metals, elastomers are complex because both their material and

geometric behavior is non-linear. Elastomers are isotropic, highly deformable, highly elastic,

and nearly incompressible. They are also hyperelastic or Green elastic, i.e., their mechanical

properties can be characterized by means of a strain energy function. Elastomers have high

elongation before fracture and a very low elastic modulus and relatively low thermal

conductivity. In addition, the mechanical properties of elastomers are highly sensitive to

strain, strain rate, strain history, temperature, composition, and manufacturing methods.

Most elastomeric parts are produced by a molding process in which a composition of raw

materials and additional additives are subjected to a controlled temperature-pressure-cycle to

produce the desired properties and geometry. Some elastomer applications such as tire or

shoe tread patterns require a complex mold to be manufactured. The disadvantage of this

method is the high cost and time-consuming process of manufacturing the mold. As an

alternative, elastomer machining is proposed in this research to reduce production costs by

eliminating the molding process, which would be especially useful for manufacturing

prototype parts and other applications requiring a complex shape. Potential applications of

machined elastomers include prototype tire and footwear tread patterns, scrap tire recycling,

(25)

With the rapid advancement of computational power, numerical techniques have been

developed to analyze complex engineering problems. Among available numerical methods,

the Finite Element Method (FEM) is considered to be one of the most powerful, accurate and

reliable computational tools. The FEM has been used extensively to analyze manufacturing

processes such as machining and forming of more traditional engineering metallic materials

such as steel and aluminum alloys (Strenkowski and Carroll, 1985), (Strenkowski and Moon,

1990), (Strenkowski and Lin, 1996), (Shih, 1995), (Shih, 1996). However, very little

research has been undertaken to investigate the machining and chip formation process of

elastomers with the finite element method.

The roughness of a machined elastomer surface is highly dependent on the type of chip

produced during machining. Discontinuous chip formation is associated with a rough surface

finish of the workpiece, which results in inaccuracy in the final machined workpiece

dimensions. In this research, wedge indentation was investigated because indentation can be

considered to be the incipient stage of chip formation in a machining process.

Very little research on elastomer machining has been performed because of the complexity of

elastomers and the machining process itself. In this research, better understanding of

machining of elastomers was achieved through the development of wedge indentation

models and orthogonal cutting tests. The wedge indentation models were used to investigate

chip formation during incipient metal cutting prior to material separation. Even though

wedge indentation does not incorporate chip formation, it can lead to a better understanding

(26)

1.1 Scope and Objective of Dissertation

The main objective of this research is to gain a better understanding of the machining

conditions that result in improved machinability of elastomers. A series of half-wedge

indentation models were developed and orthogonal cutting tests were conducted to achieve

this objective. Simulations of half-wedge indentation in elastomers were conducted to

identify machining conditions that lead to continuous chip formation and good surface finish.

These results were confirmed with orthogonal cutting tests for tools with various rake angles

under different feed speed conditions.

The following tasks were undertaken in this research:

1. An extensive literature review of wedge indentation and machining of elastomers was

conducted. The material characteristics of elastomers and the associated constitutive

equations were reviewed. The mechanics of machining for both metals and elastomers

was also reviewed, as well as the use of the finite element method for simulating

indentation and machining processes.

2. A series of orthogonal machining experiments of elastomers was conducted using a

conventional lathe under various cutting conditions. The resulting chip morphology and

machined surface finish were observed and recorded and the resulting forces were

(27)

3. The effect of various machining parameters such as workpiece fixturing, tool rake

angle, and feed speed on the generated chip geometry and machined surface quality of

the elastomeric workpiece was investigated. These results provided useful information

for identifying machining conditions that lead to continuous chip formation and a

smooth surface finish.

4.

Two-dimensional finite element models of half-wedge indentation of elastomers were

developed using ABAQUS, a commercial finite element analysis program. These

models were used to identify the initial stage of material separation which subsequently

leads to incipient chip formation in machining. These models were used to characterize

the workpiece material response for ultimately identifying the cutting conditions for

which indentation transitions to cutting and chip formation. A preliminary model of

(28)

1.2 Overview of Dissertation

The following chapter presents a literature review of the material characteristics of

elastomers and the associated constitutive equations. The review also includes past research

regarding the indentation of metals and elastomers and the mechanics of the machining

process. Background on the use of the finite element method for simulating cutting

processes is also described. In Chapter 3, orthogonal cutting tests for an elastomer are

described. The chip morphology and corresponding machining forces under various cutting

conditions are discussed. The effects of various machining parameters such as the tool rake

angle and feed speed on the machined surface finish are also examined in this chapter.

Chapter 4 discusses the development of half-wedge indentation models for elastomers. For

comparison, wedge indentation and orthogonal cutting models for AISI 4340 steel are also

described. Background on the use of the non-linear finite element technique is given for

simulating the indentation of elastomers. Results from the elastomer half-wedge indentation

and metal cutting models are discussed in Chapter 5. Preliminary results of an elastomer

orthogonal cutting simulation are also presented. Conclusions and recommendations for

(29)

2. LITERATURE REVIEW

2.1 General Description of Elastomers

An elastomer can be defined as, “a macromolecular material, which, at room temperature, is

capable of recovering substantially in shape and size after removal of a deforming force.”

(ASTM, 1999). Rubber is also defined by the same standard as “a material that is capable of

recovering from large deformations quickly and forcibly and can be, or already modified to a

state in which it is essentially insoluble (but can swell) in boiling solvent, such as benzene,

methyl ethylketone, or ethanol-toluene azeotrope. A rubber in its modified state, free of

diluents, retracts within one minute to less than 1.5 times its original length after being

stretched at room temperature to twice its length and held for one minute before release.”

Therefore, it is noted that all rubbers are elastomers; however, not all elastomers are rubbers.

Some types of plastics can be also considered to be elastomers, according to this definition.

Among a variety of plants containing rubber, the rubber tropical trees (Hevea brasiliensis)

growing mainly in Southeast Asia and Africa are the major source employed for commercial

rubber. Nevertheless, any synthetic material such as neoprene, nitrile, and styrene butadiene

are categorized with natural rubbers since their characteristics can be classified as rubbers

(Shalaby and Schwartz, 1985).

The high elasticity of elastomers that allows for large elastic deformation is certainly one of

the most remarkable characteristics. Elastomers have a very low elastic modulus and a high

(30)

hysteresis, which contributes to their energy absorption capability. A crosslinked strip,

extended to many times of its original length when released, will return to that original length

showing very little or no permanent deformation as a result of the extension. This is in

contrast to the behavior of ductile metals that can undergo large deformation without fracture

but do not return to their original shape upon removal of the load (Aklonis, MacKnight et al.,

1972).

The deformation of elastomers can be considered to be non-linear in terms of both their

material and geometric behavior. Elastomers are isotropic, highly deformable, highly elastic,

nearly incompressible, hyperelastic or Green elastic, i.e., their mechanical properties can be

characterized by means of a strain energy function (Ogden, 1997). In addition, the

mechanical behavior of elastomers is very complex due to their sensitivity to strain,

strain-rate, deformation history, temperature, and the fabrication process. Table 2.1 illustrates a

(31)

Table 2.1 Comparison of properties of elastomers with other materials (Strenkowski, Shih et al., 2002).

Material Modulus of Elasticity (GPa) Poisson’s ratio Ultimate stress (MPa) % of elongation to fracture Thermal conductivity (W/m K) Elastomers 0.0007 – 0.004 0.47 – 0.5 7 – 20 100 – 800 0.13 – 0.16 Aluminum alloy 70 – 79 0.33 100 – 550 1 – 45 177 – 237 Steel,

high-strength

190 – 210 0.27 – 0.3 550 – 1200 5 – 25 35 – 60

Steel, spring 190 – 210 0.27 – 0.3 700 – 1900 3 – 15 40 – 50 Plastic, Nylon 2.1 – 3.4 0.4 40 – 80 20 – 100 0.3 Plastic,

Polyethylene

0.7 – 1.4 0.4 7 – 28 15 – 300 0.4

Note that the modulus of elasticity, the ultimate stress and the thermal conductivity of

elastomers are very low compared to most metals. The Poisson’s ratio of elastomers reaches

0.5 due to its incompressibility.

The deformation of rubbers began to receive serious attention from engineers and scientists

in the 1940’s. Rivlin was one of the leading researchers who worked with the British Rubber

Producers’ Association. His major contribution included an introduction of the theory of finite elastic strain or hyperelasticity as applied to rubber. Subsequently, several other

researchers including Ogden, Valanis, and Yeoh developed various constitutive relationships

for rubber materials.

In general, the elasticity of elastomers can be studied in three categories; namely, a

thermodynamic approach, a statistical approach and a phenomenological treatment. The first

thermodynamic treatment is only concerned with the macroscopic behavior of the material,

(32)

underlie elastomeric molecular dynamics. The phenomenological approach, which is widely

used today, provides a description of the elasticity for large deformation. As its name

implies, the phenomenological approach is concerned only with observed behavior, not with

the mechanism responsible for that behavior. Its primary aim is to provide a means of

calculating the relationships between the state of strain in a deformed body and the applied

forces. A secondary aim is to find a general description of material properties in terns of the

elastic energy stored in the system (Treloar, 1958), (Aklonis, MacKnight et al., 1972). The

elastic energy stored after the deformation in a system is a unique state function of strain and

it is independent of the strain path. The state of strain is defined by three principal stretch

ratios, denoted by λ1, λ2, and λ3 (Ogden, 1997). The strain energy function used in the

statistical and phenomenological approaches is discussed in the following section.

2.2 Analysis of the Strain Energy Function for Elastomers

2.2.1 A Statistical Approach

The statistical approach is based on the concept that valcanized rubber is an assembly of

long-chain molecules, linked together with a relatively small number of points to form an

irregular three-dimensional network. Under the assumption that there is no change in the

internal energy of the system during deformation, the free energy per unit volume or the

strain energy density (W ) can be written in the form (Treloar, 1958):

W = 2 1

(33)

Where λ1, λ2 , λ3 are the extension ratios along three mutually perpendicular axes, N is the

number of chains per unit volume, k is the Boltzmann constant, and T is the absolute

temperature. It is more convenient in practice to write (2.1) as:

W = 2 1

G (λ21 + λ22 + λ23 – 3) (2.2)

Where G is the shear modulus or modulus of rigidity, and G = NkT. This relation is also referred as the Neo-Hookean function.

2.2.2 A Phenomenological Approach

Based on the assumption that rubber is isotropic and incompressible, Mooney developed an

expression of the strain energy function as (Mooney, 1940):

W = C1 (λ21 + λ22 + λ23 – 3) + C2 (λ-21 + λ-22 + λ-23– 3) (2.3)

Where C1 and C2 are material constants that are experimentally determined.

Rivlin introduced an expression of all possible forms of the stored energy, W, for isotropic

and incompressible materials, as a function of three quantities, I1, I2, and I3 called strain invariants which can be defined as (Rivin, 1948):

I1 = λ21 + λ22 + λ23 (2.4)

I2 = λ21λ22 + λ22 λ23 + λ23λ21 (2.5)

(34)

The assumption of incompressibility, i.e. the volume of material is constant in all states of

stress, provides the relation:

I3 = λ21 λ22 λ23 = 1 (2.7)

Therefore, I3, is not a function of strain, and it can be concluded that, for an incompressible

elastic material which is isotropic, the stored energy, W, is defined as a function of two independent variables, I1 and I2 and it may be expressed as the sum of a series of terms as follows:

W =

∑

∞

= =0,j 0

i

Cij(I1 – 3)i(I2- 3)j (2.8)

This equation is called the Rivlin function in general form. Note that this expression is

arbitrarily written in terms of I1 and I2 so that W vanishes at zero strain (I1 = I2 = 3) by means

of equations (2.4) – (2.5).

It can also be seen that the Mooney strain energy function in (2.3) is a special case of

equation (2.8) where the first order relationship of the Rivlin function is considered, and it

can be rewritten as:

W = C1 (I1 – 3) + C2 (I2 – 3) (2.9)

(35)

Yeoh also defined the strain energy function in the form of (Yeoh, 1993):

W = C1 (I1 – 3) + C2 (I1 – 3)2 + C3 (I1 – 3)3 (2.10)

Valanis and Landel proposed another model which describes the strain energy function as the

sum of three separate functions of the three principal stretch ratios as (Valanis and Landel,

1967):

W = w(λ1) + w(λ2) + w(λ3) (2.11)

Among the forms of the function w(λi), the Ogden-Tschoegl model (Ogden, 1997), (Blatz,

Sharda et al., 1974) is one of the most widely used today. It is defined as:

W =

∑

=

m

k 1

µk (λ1αk + λ2αk + λ3αk – 3) / αk (2.12)

where αk (k = 1,..,m) is a real number and m is a positive integer. Both µk and αk are

empirical material constants. Note that when m =2, αk = ±2, and µ1 = ±2Ck, k = 1, 2, the

strain energy function in equation (2.12) becomes the Mooney-Rivlin strain energy function

shown in equation (2.9). When m =1, α1 = 2, and µ1 = G, the strain energy function becomes

the Neo-Hookean function shown in equation (2.2).

The Ogden-Tschoegl model has been applied to various tests. For simple tension using a

3-term function, it can describe a strain range up to 700% with small deviation compared to

(36)

2.3 Analysis of a Wedge Indentation Process

The machining of metals has been successfully studied as a transition from wedge

indentation. The mechanism of material separation in wedge indentation leads to incipient

chip formation in metal cutting. Therefore, a review of wedge indentation of metals is given

next. In addition, previous research in wedge indentation of elastomers was also surveyed.

2.3.1 Wedge Indentation of Metals

In general, plasticity solutions of wedge indentation of metals are classified into two

categories (Lockett, 1963). The first category is incipient deformation. A body is subjected

to a given set of boundary conditions and then the initial stress and velocity fields are

calculated. Due to the time dependence of loading and boundary conditions, such solutions

are only applied immediately after load application. An example of work in this category

was conducted by Shield (Shield, 1955). In this work, the indentation of a semi-finite body

by a rigid circular flat punch was analyzed to obtain the stress fields.

The second category is a class of problems for which plane strain conditions are assumed and

the slip-line field is constructed from basic elements involving straight lines and arcs. As a

result, the analytical solutions for the stress and velocity fields can be found. A theoretical

solution for the deformation caused by a rigid lubricated wedge indentation of a plastic

metallic material (Hill, Lee et al., 1947) was developed using this method. In this study, the

metal was considered to have a distinct yield point and the rate of work-hardening was

neglected. The effect of specimen size in the measurement of a hardness test using flat dies

(37)

of wedge indentation as a precursor to cutting. In this work, a flat strip of metal was cut in

two parts using a wedged-shaped cutting tool (Hill, 1953). The width/thickness ratio of the

strip was large enough so that the deformation could be treated as plane strain. As the strip

was indented, the plastic zone extended to the base until the sideways force exerted by the

wedge reached a value such that a plastic zone formed through the remaining thickness of

material. As a result, separation of the material was initiated.

The cutting of round wires of copper and steel with flat and wedge-shaped tools has also

been carried out (Johnson, 1958). Plane strain conditions were assumed in this work.

Load-displacement relationships were determined. The effects of wire diameter and wedge angle

on the maximum cutting required load were also determined.

Slip line theory has been used to analyze the indentation of a sharp edge punch to determine

the mechanism of chip formation during incipient cutting with a negative rake angle tool

(Okushima and Hitomi, 1963). Good correlation between developed expressions of tool/chip

contact length, chip thickness, and cutting forces with experimental measurements was

obtained

A study of incipient chip formation during orthogonal cutting of brass was carried out

(Weinmann and Von Turkovich, 1971). Various cutting tools with negative, zero and

positive degree rake angles were studied. A hardness and recrystallization investigation

revealed that a maximum shear zone originated from the cutting edge vicinity and extended

(38)

(half-wedge and full-(half-wedge configurations) and distance between the indentation point and the

free edge were included in the study (Weinmann, 1977).

Wedge indentation of brass was also used to evaluate the chip separation criterion for

incipient cutting (Strenkowski and Mitchum, 1987). In this work, the beginning of incipient

cutting was determined to occur when the effective stress contour of the yield zone extended

completely from the tool edge to the outside edge of the workpiece. At this point, the

effective strain at the node adjacent to the tool edge was identified as a separation criterion

for subsequent chip formation.

Using the finite element method, Bhattacharya and Nix simulated the load-depth relationship

of a sub-micrometer indentation test for nickel, silicon, and aluminum with a rigid cone

indenter (Bhattacharya and Nix, 1988). The hardness and the elastic modulus were obtained

from the loading and unloading portions of the curve. Later, simulation of a frictionless

wedge indenting a copper strip under plane strain conditions was presented (Jayadevan and

Narasimhan, 1995).

Simulation of wedge indentation with various cone angles was also carried out to determine

the stress-strain relationships of nickel and copper through a semi-inverse method (Dicarlo,

Yang et al., 2002). In this investigation, power law hardening and the ratios of yield strength

to Young’s modulus were included in a finite element model of the indentation process.

Results from the model were used to correlate the material properties with variations in

(39)

application of both the finite element and boundary element methods for simulating the

indentation process can be found in a paper published by Mackerle (Mackerle, 2001).

2.3.2 Wedge Indentation of Elastomers

The indentation of thin rubber sheets with various thicknesses by a rigid cylindrical indenter

was carried out by Waters (Waters, 1965). Using a classical elasticity solution, relationships

of the total load applied on a die, a die radius, and a sheet thickness were established. An

analysis of a durometer indentation of elastomers was carried out (Briscoe and Sebastian,

1993). The theory of elasticity was used to derive relationships between the Shore hardness

number and Young’s modulus. A comparison of the standard indentation hardness test

procedures, namely, International Rubber Hardness (IRH) and the Shore A Hardness with the

corresponding elastic modulus was also developed.

Rubber indentation with various indenter geometries such as a cone, sphere, cylinder, has

been examined (Briscoe, Sebastian et al., 1994). The influence of the indenter geometry on

the elastic modulus of rubber was also investigated in this work. The effects of the indenter

size, depth, specimen size, and rubber material properties have also been studied both

empirically and numerically (Karduna, Halperin et al., 1997).

The finite element technique with adaptive meshing was employed for plain strain

(40)

tensile tests have been performed to determine material parameters used in the

Mooney-Rivlin, Yeoh, and Ogden material models (Jerrams, Kaya et al., 1998).

Based on this review, the major goal of past research has been to determine the material

properties of elastomers by characterizing their stress-strain behavior. In contrast, a major

goal of the present research is to investigate indentation of an elastomer material to the point

at which separation is likely to occur as a precursor to incipient cutting.

The failure of elastomer components has also been investigated by several researchers. The

fracture mechanics approach is widely adopted to study the strength of elastomer. The

concept originally proposed (Rivlin and Thomas, 1953) assumed that failure occurred locally

by crack growth resulting from imperfections such as small cracks present in the material. In

this work, the critical amount of energy required to advance a crack per unit area was

expressed by: l

A

W

T













∂

−

=

_(2.13)

where T = strain energy release rate or tearing energy

W = total strain energy stored in the material

A

= area of one fracture surface of the crack

l

= constant length for which the applied external force does

no work.

(41)

The tearing energy concept was used as a fracture criterion to predict the critical load for

crack propagation (Pidaparti and Pontula, 1995). In this work, the fracture criterion is

defined by:

c

T

≥

(2.14)

where

T

c = critical tearing energy obtained empirically.

As previously described, the constitutive behavior of hyperelastic materials can be

characterized in terms of the strain energy density. There is evidence from previous research

that the strain energy is a critical parameter for predicting the rupture failure of elastomers.

Therefore, in addition to several stress measures such as the maximum principal stress in the

workpiece under the indenter tip, the strain energy density will be one of the variables to be

studied for predicting incipient chip formation in wedge indentation of elastomers.

2.4 Analysis of a Machining Process

Machining involves the removal of layers from a workpiece in the form of a chip by action of

a wedge shaped cutting tool. In general, chip formation can be categorized based on the

cutting geometry and the deformation within the primary zone. For cutting geometry, the

(42)

Chip

Cutting edge inclination angle Workpiece motion

Workpiece

Tool

Orthogonal Cutting Oblique Cutting

Chip

Workpiece

Tool

Workpiece motion

Figure 2.1 Comparison between orthogonal and oblique cutting.

Workpiece Chip thickness

Depth of cut

Shear angle

Cutting direction

Chip

Rake angle Tool

Clearance angle

(43)

In orthogonal cutting, the cutting edge of the tool is perpendicular to the direction of relative

workpiece-tool motion and the velocity of the workpiece material is orthogonal to the tool

cutting edge. Orthogonal cutting has been extensively used by several investigators to

simplify the analysis of the cutting process. In practice, very few machining processes can

be classified as orthogonal.

In contrast, oblique cutting refers to cutting when the cutting edge of the tool is oblique or at

an angle to the movement of the relative workpiece-tool direction. The velocity of the

workpiece material is inclined to the cutting edge at an inclination angle. Most machining

processes such as helical end milling and drilling are classified as oblique cutting. Figure 2.1

shows both orthogonal and oblique cutting schemes, and the nomenclature of orthogonal

cutting is illustrated in Figure 2.2.

Most analyses of the mechanics of cutting have been applied to a metal workpiece. Very

little research has been conducted on the machining of elastomers. Kolman and Rudziejewski

presented the effects of cutting parameters on the machined surface roughness in turning of

rubber (Kolman and Rudziejewski, 1968). A variety of carbide end mills of various sizes

and helix angles were used to mill grooves in three types of elastomers; H-NBR, Norbornone

rubber, and silicone rubber at various cutting speeds (Jin and Murakawa, 1998). At high

helix angle cutters and high speeds, smoother machined surfaces and lower forces were

discovered. Recently, end milling and turning experiments on elastomers were conducted

(Strenkowski, Shih et al., 2002). In this research, cutting tools with various diameters,

materials and geometries were used in end milling for several cutting conditions with varying

(44)

cutting conditions were major factors in achieving smooth machined surfaces. In addition,

high-speed steel (HSS) tools with different rake angles were used to examine chip formation

under orthogonal cutting conditions in a turning process. The results showed that both

discontinuous and continuous chips occurred for the range of cutting conditions studied. Feed

and tool rake angle were found to have a significant effect on the type of chips produced

using turning.

End milling of elastomers has also been investigated (Shih, Lewis et al [a]., to be published),

(Shih, Lewis et al [b]., to be published). In this work, the design of the workpiece fixture

was found to be a critical factor in achieving good surface finish in end milling of elastomers

because of the low elastic modulus. Finite element analysis was used to analyze the stiffness

of the elastomer workpiece to achieve sufficient fixture rigidity. The chip morphology and

cutting forces in end milling were also analyzed. Chip classification based on size and

morphology was examined using Scanning Electron Microscopy (SEM) micrographs.

Cutting force components were recorded using both solid carbon dioxide cooled elastomers

and elastomers at room temperature. The maximum uncut chip thickness at the average peak

cutting force was identified for various spindle speeds. The possibility of adiabatic shear

band formation in serrated chip formation due to low thermal conductivity of elastomers was

also suggested. This research provided valuable background for investigating the mechanics

of elastomer machining.

Considering that there is a large percentage of elongation before failure and that the elastic

(45)

of metals and elastomers. For instance, fixture design of the workpiece is very critical so

that adequate structural support is provided. In addition, the cutting forces for elastomer

machining are much lower than those in conventional metal cutting. These factors are

important considerations when investigating the machining of elastomers.

With the advances in computational technology in recent decades, finite element analysis

(FEA) has become a powerful technique in modern engineering practice. Metal cutting has

been successfully analyzed using FEA with promising results. Strenkowski and Carroll

developed an orthogonal cutting model using an updated-Lagrangian elasto-plastic dynamic

formulation (Strenkowski and Carroll, 1985), (Strenkowski and Carroll, 1986). In this work,

materials were considered to be elasto-plastic and strain rates were neglected. The tool was

advanced into the workpiece from the incipient stage until steady state cutting was reached.

However, the mesh was attached to the workpiece being modeled, which could cause

problems due to severe deformation of the workpiece. Also, a prescribed chip separation

criterion based on effective strain must be known in advance. The model was improved

(Strenkowski and Mitchum, 1987); however, the accuracy was limited by the size of the

elements in the vicinity of the cutting edge. Later, Strenkowski and Moon introduced an

alternative Eulerian approach for simulating a flat tool. In this study, the grid was spatially

fixed and the mesh was considered to be a control volume with a stationary grid that can be

refined as necessary at the tool cutting edge where the stress and strain gradients are the

highest (Strenkowski and Moon, 1990). The workpiece was treated as a viscoplastic material

in which the material can flow through the control volume. The model included temperature

(46)

The restriction of a predefined tool-chip contact that existed in previous models was later

modified (Athavale and Strenkowski, 1995). Further development of the cutting model was

carried out by Hsieh (Hsieh 1998) to improve the model accuracy and capability for use in

cutting tools with large negative rake angles. In this work, a three-dimensional twist drill

model was developed in conjunction with an analytical single edge oblique cutting model

(Usui, Shirakashi, et al. 1978). The calculated drilling forces, torque and temperature

distribution in the drill body were obtained in this investigation.

A comprehensive finite element model based on an updated Lagrangian approach as

formulated in ABAQUS, a commercial finite element code, was used to simulate chip

formation in an orthogonal metal cutting process (Komvopoulos and Erpenbeck, 1991). The

effects of interfacial friction on the chip-tool interface and tool wear geometry were

investigated in this study. The development of a plane strain finite element analysis for

orthogonal metal cutting simulation was also carried out by Shih (Shih, 1995), (Shih, 1996).

The cutting process was analyzed using an unbalanced force reduction method and a

sticking-sliding friction model. The effects of work material, viscoplasticity, temperature,

strain and strain-rate were included in this study. The predicted residual stress distribution

was correlated with x-ray diffraction measurements for various rake angles (Shih, 1996).

With the rapid growth of general purpose finite element programs such as ABAQUS,

modeling of the cutting process can be readily achieved. The effect of sequential cuts on the

residual stress in the machined layers was predicted using a thermo-elastic viscoplastic finite

(47)

work, chip formation, cutting forces, and temperatures were predicted for stainless steel. In

another study, a detailed procedure for modeling the metal cutting process using

ABAQUS/Standard was described (Shet and Deng, 2000). Predictions of the stress, strain,

strain-rate fields and temperatures were obtained for various various tool rake angles and

friction coefficients on the chip-tool interface. Recently, Unanue developed a cutting model

using ABAQUS/Standard with chip separation based on a critical stress criterion at the

cutting line (Unanue, Rubio et al., 2001).

Based on past work, it is evident that commercial finite element software such as ABAQUS

offers potential for simulating the metal cutting process. In this dissertation,

ABAQUS/Standard and ABAQUS/Explicit were adopted to simulate both half-wedge

indentation and orthogonal cutting of AISI 4340 steel and elastomers. A preliminary

ABAQUS/Standard model for simulating orthogonal cutting of elastomers was also

(48)

3. ELASTOMER MACHINING EXPERIMENTS

3.1 Introduction

A series of orthogonal machining tests of elastomers on a conventional lathe was conducted

to gain a better understanding of elastomer chip formation and to establish cutting conditions

necessary to achieve a good surface finish. In orthogonal cutting, chips flow along the tool

rake face in two-dimensional plane-strain deformation without lateral motion. Although

turning is a three-dimensional cutting process in general, the orthogonality assumption

greatly simplifies the geometry of the cutting process, while retaining the material response

characteristics of the tool/workpiece interaction. Therefore, orthogonal machining was

considered to be an appropriate assumption in this research. To obtain orthogonal machining

conditions, an elastomer tube with large radius and a small thickness was used so that the

cutting velocity variation along the tube thickness was negligible.

3.2 Experimental Setup

Elastomer machining experiments were conducted on a conventional lathe as shown in

Figure 3.1. Rubber tubes obtained from Lord Corporation (Batch No. CS-02-003288) were

used as specimens in this study. Each rubber tube had an outside diameter of 95.25 mm and

a wall thickness of 11.61 mm. Great care was taken to achieve sufficiently stiff fixturing of

the rubber tube. The fixture consisted of attaching the rubber tube to an aluminum mandrel

(49)

in Figure 3.1. The aluminum mandrel provided a strong structural support that constrained

the rubber workpiece against the cutting forces. In the experiments, the edge of the rubber

tube overhung approximately 38 mm from the aluminum mandrel edge so that only the

rubber was machined during the cutting tests. A pipe clamp was also used to secure the

rubber specimen to the aluminum mandrel. The aluminum mandrel with the attached rubber

workpiece was secured in the lathe chuck as shown in Figure 3.2.

(50)

Figure 3.2 Side view of workpiece in the lathe chuck for the elastomer machining experiments.

Elastomer machining experiments were performed on a 5 horsepower Cincinnati Hydroshift

lathe. Various cutting tool geometry and cutting conditions were used to study the effects on

chip morphology, machined surfaces, and cutting forces. A set of 12.7 mm square solid high

speed steel (HSS) cutting tool blanks were ground to a sharp edge with a clearance angle of 5

degrees and rake angles of 10, 30, and 50 degrees. The nomenclature of the cutting tool

angles is shown in Figure 3.3. Two feed speeds of 0.0635 and 0.2540 mm/rev and a cutting

speed of 2.50 m/s were used in these tests. The tests were conducted at room temperature

under dry conditions. Elastomer chips were collected in each cutting experiment and

photographs of the chips and the machined surfaces of rubber were recorded for further

(51)

Rake angle

Clearance angle